Different benchmarks:

> Artificial Analysis Agentic Index: Represents the weighted average of agentic capabilities benchmarks in the Artificial Analysis Intelligence Index (GDPval-AA v2, Tau³-Banking)

> Artificial Analysis Coding Agent Index v1.3 incorporates 3 benchmarks: DeepSWE, Terminal-Bench v2, and SWE-Atlas-QnA

Qwen3.8 Max is 55.4 on the Agentic Index but hasn't been tested for the Coding Agent Index.

Indeed, and this Qwen 3.8 max specific page:

https://artificialanalysis.ai/models/qwen3-8-max

Doesn't have the claim either. Clickbait?

According to those graphs, Grok 4.5 appears to be the most cost-effective model.

They didn't run all benchmarks. It's the best in AA agentic index (GDPval-AA v2, ³-Banking) but not coding index (DeepSWE which is missing, Terminal-Bench v2.1 they have 81% vs 90% for Sol, SWE-Atlas-QnA missing).

[delayed]

What type of agentic tasks are you using it for (eg how complex)?

How are you running it on a Strix Halo? The weights aren't out yet, are they?

It's not enough that it's better?

Many providers will host it and will compete on price. It also can't easily be taken away because one company (or one government) decides they don't want it around any more. People can fine-tune it for particular workloads.

For one thing, providers of open models can't arbitrarily increase their prices without facing competition.

How CLI are you guys using for qwen and kimi?

It’s because they’re doing some sort of combined score of intelligence, speed and cost. On pure intelligence it doesn’t even show up in the top 10.

Depends on what you want to do. Some task specific models can be trained with a few ten or hundred thousand training examples so you can use a bigger model to produce synthetic training examples and then fine tune a smaller student model. I think that's the usual process. Whether you'd get acceptable performance this way depends, as mentioned, on what you're trying to do and what you'd consider acceptable.

once you are able to get the full probability distributions per token you can distill it on specific domains. distilling without that isn't generally a good idea unless you have invested millions in the requisite infrastructure.

Almost surely. Apple is extremely well positioned to take advantage of this over the next decade.

I'm dumbfounded to see Opus 5 making SO MANY mistakes in coding simple stuff. Most times, Fable 5 comes out to be cheaper because it nails so many things much quicker than Opus 5.

Sent to solve one task, came back with half of it solved and 2 more problems.

"As you requested, I've finished task X. Honestly, task X turned out to require task Y, which I haven't actually done. Task Y is the next step if you'd like to continue along this route."

> less knowledge, more intelligence

People produce such models by over-RL-ing smaller models on math and coding tasks. I've found the results capable of neither innovative work nor thinking outside the box. They're straight-A students raised by tiger moments who never let them play freely for hours in the dirt.

Perhaps you could say such models are skilled --- but intelligent? Not by my measure.